In vivo enhancer screening and genome engineering of the 9p21 diabetes locus in disease relevant human cells will potentially discover novel mechanisms of inherited predisposition. The pancreatic beta cell plays an important role in the course of type 2 diabetes. Autopsy series suggest that beta cell mass is decreased before disease onset. Therefore, a fundamental advance for the understanding and prevention of diabetes would be the identification of predominant genetic variants that contribute to decreased insulin in those at risk for developing the disorder. Multiple genomic locations harbor sequence variants that correlate with inherited risk for diabetes but with no known mechanism. Mounting evidence implicates CDKN2A as a master regulator of pancreatic beta-cell specification. Furthermore, non-coding polymorphisms in the CKN2A/CDKN2B locus (9p21) reproducibly correlate with the disease. Herein, we propose a novel approach aimed to (1) use molecular biochemical methods to discover in vivo enhancer elements within the 9p21 diabetes locus, (2) test disease causation by the targeted genomic deletion of the 9p21 diabetes-associated element using novel homologous recombination techniques (3) explore the impact the deletion has on pancreatic beta-cell lineage differentiation and local gene expression. The strength of this three-pronged approach is that it is an end-to-end in vivo investigation of human biological regulation at non-coding loci implicated in type 2 diabetes predisposition. This proposal, though ambitious, will provide a powerful framework for the direct testing of genetic causation on diabetes relevant endpoints. The success of any of these aims alone or in combination has the potential to make meaningful insights into previously unknown disease mechanisms as well as highlight potential sequence variants that could predict disease risk. Furthermore, if successful, this approach could pioneer a new methodology for relating the contribution of non- coding risk-loci to the underlying mechanisms of all complex disease/traits.